Verified65% confidenceBenchmarkExact time
In the 12-question hands-on test, Gemini 4 Argon's record against Claude Opus 5.5 was 0 wins, 5 draws and 7 losses, and against GPT-6.1 Sol was 2 wins, 4 draws and 6 losses, ranking last among the three.
3
Sources
65%
Confidence
Short-term (~14 days)
Relevance
10/14/2026
First Seen
Valid until: 10/28/2026
Sources
Related Entities
Related Claims
UnverifiedIn the legal work test, Gemini 4 Argon scored 19.6 and Claude Opus 5.5 scored 3.8, making Argon more than five times Opus.78% similarVerifiedOn the DeepSWE v1.1 (coding) benchmark, Gemini 4 Argon scores 77.9%, Claude Opus 5.5 scores 74.2%, and GPT-6 Astra scores 74.1%.76% similarUnverifiedIn the machine learning engineering test, Claude Opus 5.5 scored 49.3 and Gemini 4 Argon scored 45.3.74% similarUnverifiedIn project three (Lakebed product audit), Gemini 4 Argon scored only 2.25/10, with 56 of 78 findings being wrong, a failure rate of about 70%; while 6.1 Sol scored 9/10.73% similarPartially VerifiedOn the Terminal Bench terminal task test, Gemini 4 Argon scored 57.4%, below Claude Opus 5.5's 66.4%.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030276API
curl https://kongchang.com/api/v1/knowledge/claims/1030276MCP
get_claim(id=1030276)